KR102548104B1 - Method, computer program, and device for generating training dataset to identify hit location of dart pin - Google Patents

Abstract

The present invention relates to a method for generating a training dataset for dart pin location identification, a computer program, and an apparatus thereof. According to some embodiments of the present disclosure, disclosed is the computer program stored on a computer-readable storage medium. The computer program, when executed on equal to or more than one processor of the computing apparatus, provides the method for generating the training dataset for the dart pin location identification. The method may comprise the following steps of: obtaining a dart target shooting training image in which at least one dart pin hits one segment of a dart target (each of a plurality of segments included in the dart target includes a predetermined number of bits); and assigning a label corresponding to the obtained dart target shooting training image.

Description

Method, computer program, and device for generating training data set for identifying dart pin location

The present disclosure relates to image processing using a computing device, and more particularly, to a method, computer program, and device for generating a training dataset for identifying hit locations of dart pins.

In general, darts mean 'small arrows' and are a game in which arrow-shaped darts are thrown at centrifugal targets with numbers written on them to score points. The dart game has an advantage that anyone can always enjoy it if an arrowhead-shaped dart and a dart game device are provided. Recently, with the development of various game methods and the arrangement of scoring methods, it has developed into a world-class leisure game, and men and women of all ages are enjoying the game conveniently.

The dart game device may identify the hit position of the dart pin hit on the dart target by analyzing a sensor installed on the dart target or an image of the dart target. The image analysis method needs to process images taken from cameras in various locations. The image analysis method using the rule-based algorithm has a problem in that it is difficult to secure accuracy due to noise in that the dart game device is in various installation environments. The deep learning-based image analysis method has problems such as difficulty in securing a sufficient amount of dataset and implementation of a hit location processing algorithm of a plurality of dart pins.

Korean Registered Patent No. 10-2136813

The present disclosure has been made in response to the above background art, and is to provide a method, computer program, and apparatus for generating a training data set for identifying hit locations of dart pins.

The technical problems of the present disclosure are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the following description.

According to some embodiments of the present disclosure for solving the above problems, a computer program stored in a computer readable storage medium is disclosed. The computer program, when executed on one or more processors of a computing device, provides a method for generating a training dataset for identifying dart pin locations, the method comprising: hitting at least one dart pin on one segment of a dart target. obtaining one dart target shooting training image, each of a plurality of segments included in the dart target including a predetermined number of bits; and assigning a label corresponding to the acquired dart target shooting training image. can include

Alternatively, the dart target shooting training image includes at least two images corresponding to one label, and the at least two images corresponding to the one label are directed to at least two cameras respectively positioned in a predetermined direction. can be created by

Alternatively, acquiring a dart target shooting training image in which at least one dart pin hits on one segment of the dart target may include: a first dart on a first bit included in a first segment of the dart target; obtaining a first dart target shooting training image hit by a pin; and obtaining a second dart target shooting training image hit by the first dart pin on a second bit included in the first segment of the dart target. can include

Alternatively, acquiring a dart target shooting training image in which at least one dart pin hits on one segment of the dart target may include: the first bit on a third bit included in a second segment of the dart target; obtaining a third dart target shooting training image hit by the dart pin; may further include.

Alternatively, acquiring a dart target shooting training image in which at least one dart pin hits on one segment of the dart target may include: the first bit included in the first segment of the dart target; obtaining a fourth dart target shooting training image in which one dart pin hits and a second dart pin hits the second bit included in the first segment of the dart target; may further include.

Alternatively, the first dart pin hits the first bit included in the first segment of the dart target and the second dart pin hits the fourth bit included in the first segment of the dart target. acquiring a fifth dart target shooting training image; may further include.

Alternatively, acquiring a dart target shooting training image in which at least one dart pin hits on one segment of the dart target may include: the first bit included in the first segment of the dart target; obtaining a sixth dart target shooting training image in which one dart pin hits and a second dart pin hits a third bit included in a second segment of the dart target; may further include.

Alternatively, the first dart pin hits the first bit included in the first segment of the dart target and the second dart pin hits the fifth bit included in the second segment of the dart target. obtaining a hit 7th dart target shooting training image; may further include.

Alternatively, acquiring a first dart target shooting training image hit on a first bit included in a first segment of the dart target may include: acquiring the first dart target shooting training image under a first shooting environment condition; doing; and acquiring the first dart target shooting training image under a second shooting environment condition. can include

Alternatively, the label may be determined based on the location of the segment hit by the dart pin.

Alternatively, the label may be determined based on the segment location and the bit location on the one segment.

According to some embodiments of the present disclosure for solving the above problems, a method for generating a training dataset performed by a computing device is disclosed. The method includes: acquiring a dart target shooting training image in which at least one dart pin hits a segment of a dart target; and assigning a label corresponding to the acquired dart target shooting training image. can include

According to some embodiments of the present disclosure to solve the above problems, a computing device is disclosed. The computing device includes: a memory including computer-executable components; a processor executing the following computer-executable components stored in a memory; The processor may acquire a dart target shooting training image in which at least one dart pin hits on one segment of the dart target, and assign a label corresponding to the acquired dart target shooting training image. there is.

The technical solutions obtainable in the present disclosure are not limited to the above-mentioned solutions, and other solutions not mentioned will become clear to those skilled in the art from the description below. You will be able to understand.

According to some embodiments of the present disclosure, a dart game device capable of arousing the interest of dart game users may be provided.

Effects obtainable in the present disclosure are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below. .

Various aspects are now described with reference to the drawings, wherein like reference numbers are used to collectively refer to like elements. In the following embodiments, for explanation purposes, numerous specific details are set forth in order to provide a thorough understanding of one or more aspects. However, it will be apparent that such aspect(s) may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to facilitate describing one or more aspects.
1 is a block diagram illustrating an example of a dart game system according to some embodiments of the present disclosure.
2 is a block diagram illustrating an example of a dart game device according to some embodiments of the present disclosure.
3 is another diagram for explaining an example of a dart game device according to some embodiments of the present disclosure.
4 is a flowchart for explaining an example of a method of identifying a dart pin hit position performed by a dart game device according to some embodiments of the present disclosure.
5 is another flowchart for explaining an example of a method of identifying a dart pin hit location when performed by a dart game device according to some embodiments of the present disclosure.
6 is a diagram for explaining a dart target shooting image.
7 is a diagram for explaining an exemplary dart pin location identification model according to some embodiments of the present disclosure.
8 is a flowchart illustrating an example of a method of generating a training dataset for a dart pin position identification method performed by a computing device according to some embodiments of the present disclosure.
9 is a schematic diagram illustrating a network function according to some embodiments of the present disclosure.
10 is a simplified and general schematic diagram of an exemplary computing environment in which embodiments of the invention may be implemented.

Various embodiments and/or aspects are now disclosed with reference to the drawings. In the following description, for purposes of explanation, numerous specific details are set forth in order to facilitate a general understanding of one or more aspects. However, it will also be appreciated by those skilled in the art that such aspect(s) may be practiced without these specific details. The following description and accompanying drawings describe in detail certain illustrative aspects of one or more aspects. However, these aspects are exemplary and some of the various methods in principle of the various aspects may be used, and the described descriptions are intended to include all such aspects and their equivalents. Specifically, “embodiment,” “example,” “aspect,” “exemplary,” etc., as used herein, is not to be construed as indicating that any aspect or design described is superior to or advantageous over other aspects or designs. Maybe not.

Hereinafter, the same reference numerals are given to the same or similar components regardless of reference numerals, and overlapping descriptions thereof will be omitted. In addition, in describing the embodiments disclosed in this specification, if it is determined that a detailed description of a related known technology may obscure the gist of the embodiment disclosed in this specification, the detailed description thereof will be omitted. In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in this specification, and the technical ideas disclosed in this specification are not limited by the accompanying drawings.

Although first, second, etc. are used to describe various elements or components, these elements or components are not limited by these terms, of course. These terms are only used to distinguish one element or component from another. Accordingly, it goes without saying that the first element or component mentioned below may also be the second element or component within the technical spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used in a meaning commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly specifically defined.

In addition, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or”. That is, unless otherwise specified or clear from the context, “X employs A or B” is intended to mean one of the natural inclusive substitutions. That is, X uses A; X uses B; Or, if X uses both A and B, "X uses either A or B" may apply to either of these cases. Also, the term "and/or" as used herein should be understood to refer to and include all possible combinations of one or more of the listed related items.

Also, the terms "comprises" and/or "comprising" mean that the feature and/or element is present, but excludes the presence or addition of one or more other features, elements and/or groups thereof. It should be understood that it does not. Also, unless otherwise specified or where the context clearly indicates that a singular form is indicated, the singular in this specification and claims should generally be construed to mean "one or more".

In addition, the terms "information" and "data" used herein may often be used interchangeably with each other.

When an element or layer is referred to as being "on" or "on" another element or layer, it means not only directly on the other element or layer, but also intervening other layers or other elements. Including all intervening cases. On the other hand, when a component is referred to as “directly on” or “directly on”, it indicates that no other component or layer is intervening.

The spatially relative terms "below", "beneath", "lower", "above", "upper", etc. It can be used to easily describe a component or its correlation with other components. Spatially relative terms should be understood as encompassing different orientations of elements in use or operation in addition to the orientations shown in the figures.

For example, if you flip a component that is shown in a drawing, a component described as "below" or "beneath" another component will be placed "above" the other component. can Thus, the exemplary term “below” may include directions of both below and above. Elements may also be oriented in other orientations, and thus spatially relative terms may be interpreted according to orientation.

It is understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle. It should be. On the other hand, when a component is referred to as “directly connected” or “directly connected” to another component, it should be understood that no other component exists in the middle.

The suffixes "module" and "unit" for components used in the following description are given or used interchangeably in consideration of ease of writing the specification, and do not have meanings or roles that are distinct from each other by themselves.

Objects and effects of the present disclosure, and technical configurations for achieving them will become clear with reference to embodiments described later in detail in conjunction with the accompanying drawings. In describing the present disclosure, if it is determined that a detailed description of a known function or configuration may unnecessarily obscure the gist of the present disclosure, the detailed description will be omitted. In addition, terms to be described below are terms defined in consideration of functions in the present disclosure, which may vary according to the intention or custom of a user or operator.

However, the present disclosure is not limited to the embodiments disclosed below and may be implemented in a variety of different forms. These embodiments are provided only to make this disclosure complete and to completely inform those skilled in the art of the scope of the disclosure, and the disclosure is only defined by the scope of the claims. . Therefore, the definition should be made based on the contents throughout this specification.

Hereinafter, a dart game device according to the present disclosure, a dart game system, and a dart game method using the dart game device will be described with reference to FIGS. 1 to 10 . Meanwhile, in the present disclosure, for convenience of description, the player who plays the dart game and the dart game device who plays the dart game are used interchangeably without distinguishing from each other, and the accompanying drawings make it easy to understand the embodiments disclosed herein. It should be understood that the technical idea disclosed in this specification is not limited by the accompanying drawings, and includes all changes, equivalents or substitutes included in the spirit and technical scope of the present invention.

1 is a block diagram illustrating an example of a dart game system according to some embodiments of the present disclosure.

Referring to FIG. 1 , a dart game system 10000 may include a dart game device 1000, at least one other device 2000, a dart game server 3000, and a network 400. However, the above-described components are not essential to implement the dart game system 10000, so the dart game system 10000 may have more or fewer components than the components listed above.

The dart game device 1000 may provide a dart game to players participating in the dart game online. However, the dart game device 1000 may not only provide online dart games to players, but may also provide offline dart games. Hereinafter, an example of the dart game device 1000 according to the present disclosure will be described with reference to FIGS. 2 to 10 .

The other device 2000 may be at least one of devices related to players other than the dart game device 1000 . For example, the other device 2000 may be a dart game device other than the dart game device 1000, a mobile device related to the dart game device 1000 (eg, a user's smartphone on which dart game related software is installed), a server, and the like. can

As an example, the other device 2000a may be at least one dart game device participating in a dart game provided by the dart game device 1000 or waiting to participate in the dart game.

As another example, the other device 2000b may be a mobile device for expressing information related to the dart game in which the dart game device 1000 is participating by being interworked with the dart game device 1000 to the player.

As another example, the other device 2000c may be a server for providing the same dart game to a plurality of dart game devices. The server here may be the same device as the dart game server 3000 shown in FIG. 1 or may be another physically separated server. However, it is not limited thereto.

The dart game server 3000 may include any type of computer system or computer device, such as, for example, a microprocessor, a mainframe computer, a digital processor, a portable device, or a device controller.

In the present disclosure, the dart game server 3000 may determine a dart game player to participate in an online dart game.

Specifically, the dart game server 3000 may receive first play result information from a plurality of devices participating in the first game. Also, the dart game server 3000 may determine a plurality of players to participate in the second game based on the first play result information. Here, the play result information may be information about a result of the dart game generated as players participating in the dart game proceed with the dart game. The play result information may include any type of result information generated according to the play of the dart game.

As an example, the play result information may include information about ranks related to each player's dart game result, information about scores related to each player's dart game result, information about an image of a player playing a dart game, and a dictionary representing the player. It may include at least one of information about the stored image, information about the time when each player finished the dart game, or identification information of each player who participated in the dart game. However, it is not limited thereto.

For example, the dart game server 3000 may determine a plurality of players having preset rankings as players to participate in the second dart game based on ranking information included in the first play result information. . However, it is not limited thereto.

In the present disclosure, a plurality of dart game servers 3000 may be provided. Also, each of the plurality of dart game servers may distribute at least one dart game to each of the dart game device 1000 or other device 2000 .

The dart game system 10000 composed of one dart game server 3000 may be difficult to provide a pleasant dart game to players due to limitations of the network or the like. Accordingly, in one embodiment of the present disclosure, a plurality of dart game servers 3000 may be configured. Each of the plurality of dart game servers may distribute at least one dart game session or at least one sub-session to players. However, it is not limited thereto.

The network 4000 may be configured regardless of its communication mode, such as wired and wireless, and may be configured with various communication networks such as a local area network (LAN) and a wide area network (WAN).

In the present disclosure, the network 4000 or the dart game server 3000 may be a cloud-based system. Here, the cloud-based system is a system in which users (dart game devices) share network resources, and may be a computing environment in which a user borrows as much as necessary and uses it through a network at a desired time. Such a cloud-based system is a deployment model such as public cloud, private cloud, hybrid cloud, or community cloud, or IaaS (Infrastructure as a Service), A service model such as PaaS (Platform as a Service) and SaaS (Software as a Service) may be included. However, the network 4000 or the dart game server 3000 in the present disclosure is not limited to such a cloud-based system, and the centralized or edge computing network 4000 or the dart game server 3000 may also be used in the present disclosure. It will be possible to implement according to the embodiment of the content.

According to the above configuration, the dart game system 10000 is composed of the dart game device 1000, the other device 2000, and the dart game server 3000, and provides information to a plurality of players participating in the online dart game. A real-time dart game may be provided.

Meanwhile, the dart game device 1000 or other device 2000 may include components that enable a player to play a dart game. Hereinafter, components of the dart game device 1000 according to the present disclosure will be described with reference to FIGS. 2 and 3 .

2 is a block diagram illustrating an example of a dart game device according to some embodiments of the present disclosure. 3 is another diagram for explaining an example of a dart game device according to some embodiments of the present disclosure.

Referring to FIG. 2 , the dart game device 1000 includes a control unit 100, a display unit 120, a storage unit 110, a cover unit 130, a camera unit 140, a network communication unit 150, and a sensing unit. 160, a user input unit 170, a lighting unit 180, a sound output unit 190, a dart target unit 200, and at least one of the body structure 300 may be included. However, it is not limited thereto.

The control unit 100 may normally process overall operations of the dart game device 1000. The control unit 100 provides appropriate information or functions to the user by processing signals, data, information, etc. input or output through the components of the dart game device 1000 or by driving an application program stored in the storage unit 230. or can be processed.

For example, in the case of a dart game, the controller 100 may determine a score for throwing a dart pin by a player detected through the sensing unit 160 . Also, the controller 100 may transmit the determined score to the other device 2000 through the network communication unit 150 . Also, the control unit 100 may receive play-related data including scores related to the dart game from the other device 2000 through the network communication unit 150 . Then, the controller 100 controls the display unit 120 so that information related to the dart game is displayed on the display unit 120 based on the play-related data including scores of other players and the score determined by the dart game device. can Here, the play-related data may be generated as a player participating in the dart game progresses or may be data including all information that can be used to participate in the dart game. Also, the play-related data may be received again when the dart game is changed.

For example, the play-related data includes identification information of each player who participated in the dart game, information about ranks related to the dart game of each player who participated in the dart game, information about a video of the player playing the dart game, and participation in the dart game. It may include at least one of play result data for each player's dart game, information about a pre-stored image representing the player, or information about scores related to the dart game of each player participating in the dart game.

In some embodiments of the present disclosure, the controller 100 may perform a dart pin hit location identification method. A method for identifying a dart pin hit location will be described in detail with reference to FIGS. 4 to 7 .

In some embodiments of the present disclosure, the controller 100 may perform a method of generating a training dataset for identifying a dart pin location. A method of generating a training dataset for identifying dart pin locations will be described in detail with reference to FIG. 8 .

The storage unit 110 may include a memory and/or a permanent storage medium. Memory is a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg SD or XD memory, etc.), RAM (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read-Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic disk, optical disk At least one type of storage medium may be included.

The storage unit 110 may pre-store at least one image related to the dart game. In this case, the controller 100 may determine a display image related to the play-related data from among at least one pre-stored image as play-related data of the other device 2000 is received through the network communication unit 150 . Also, the display unit 120 may display the determined display image.

In detail, the information related to the dart game displayed on the display unit 120 may be a play-related image of a player of the other device 2000 . When the network communication unit 150 directly receives an image captured through the camera unit from another device 2000, the display unit 120 may display the received image.

However, according to traffic conditions between the dart game device 1000 and other devices 2000, the network communication unit 150 may receive only play-related data not including play-related images from the other device 2000. . In this case, the controller 100 may determine a display image related to play-related data among at least one image previously stored in the storage unit 110 as a display image to be displayed on the display unit 120 .

For example, the controller 100 matches the score corresponding to the triple area by the player of the other device 2000 based on the play-related data of the other device 2000 received through the network communication unit 150. and can decide. In this case, the controller 100 may determine a display image related to the triple among a plurality of display images previously stored in the storage unit 110 as a display image to be displayed on the display unit 120 .

According to some embodiments of the present disclosure, the storage unit 110 may store a dart target shooting image or a dart target shooting training image generated through the camera unit 140 . As another example, the storage unit 110 may store a dart target shooting image or a dart target shooting training image received from the outside.

The display unit 120 may display (output) information processed by the dart game device 1000 .

For example, the display unit 120 may display information related to a dart game based on play-related data of the other device 2000 received through the network communication unit 150 . Here, the information related to the dart game may be an image related to the play of the player of the other device 2000 or an image generated based on the result information of the dart game play of the other device 2000.

The display unit 120 may be provided on the front of the dart game device 1000.

As an example, referring to (a) of FIG. 3 , the display unit 120 may be provided on the front of the dart game device 1000 to display information related to the dart game.

In the present disclosure, the display unit 120 may include a plurality of display areas. For example, the display unit 120 may include a first display area 121 and a second display area 122 . Each of the first display area 121 and the second display area 122 may be independently controlled by the controller 100 . Alternatively, each of the first display area 121 and the second display area 122 may be controlled by the controller 100 so that a single image is implemented across the first display area 121 and the second display area 122. can For example, the first display area 121 and the second display area 122 may operate as independent display modules. As another example, the first display area 121 and the second display area 122 may be a single module rather than separate independent modules.

Referring to (b) of FIG. 3 , the display unit 120 may be implemented as at least one display unit. Also, each of the at least one display unit may be independently controlled by the controller 100 .

For example, the display unit 120 is located below the dart target unit 200 and has a first display area 121 and a first display unit located above the dart target unit 200 and having a second display area ( 122) may include a second display unit. Also, each of the first display unit and the second display unit may be independently controlled by the controller 100 .

The cover unit 130 is provided adjacent to the display unit 120 to protect the display unit 120 .

For example, referring to (a) of FIG. 3 , the cover unit 130 may be positioned between a virtual surface extending upward from the throw line and the display unit 120 . In this case, the cover unit 130 is positioned adjacent to the display unit 120, thereby reducing the risk of damage to the display unit 120.

Referring back to FIG. 2 , the camera unit 140 may include one or more cameras. The image frames processed by the camera unit 140 may be stored in the storage unit 110 or transmitted to the outside through the network communication unit 150 . Two or more camera units 140 may be provided depending on the use environment.

In the present disclosure, the control unit 100 of the dart game device 1000 accepts a player's login request for the dart game device 1000 based on an image including the player photographed through the camera unit 140, or can refuse

As an example, the control unit 100 may receive an input for logging in to the dart game device 1000 from a player through the user input unit 170 . In this case, the controller 100 may control the camera unit 140 to capture an image including a player currently manipulating the user input unit 170 . The controller 100 may accept or reject the player's login request based on a comparison between face information included in pre-stored player identification information and face information recognized from a photographed image.

As another example, the controller 100 may request a user's login input through the display unit 120 . Under the control of the controller 100, the display unit 120 may output information necessary for a user to log in through camera photography or log in through the user input unit 170.

As another example, the controller 100 may allow a user to log in to the dart game device 1000 through communication with another device 2000 . In this case, the control unit 100 may control the network communication unit 150 to transmit/receive information for user authentication with the other device 2000 .

In the present disclosure, the controller 100 may compare face information included in pre-stored player identification information with face information recognized from a captured image even after the player's first log-in operation.

Specifically, the controller 100 may capture a play image through the camera unit 140 to transmit a play image related to the player of the dart game device 1000 to the other device 2000 . The controller 100 may compare facial information recognized from a newly captured play video with facial information included in identification information used at login. Based on the comparison result, the controller 100 may determine whether the current player is the same player as the player at the time of login. In this case, cheating, such as participation of a substitute player, which may occur as the dart game device 1000 and other devices 2000 participate in the dart game online can be prevented. Here, the proxy player may be a player other than the player who logs in to the dart game device 1000.

As another example, the controller 100 may determine whether the dart game player cheats by comparing a pre-stored play image including a player's play motion with a current play image of the corresponding player. For example, if there is a difference between the dart pin projection image of the player included in the current play image and the past dart pin projection image of the corresponding player by a predetermined threshold value or more, the controller 100 acts as a cheat on the current dart pin projection image. probabilities can be determined. In comparing two or more images, any deep learning algorithm related to image processing or image analysis may be used.

In some embodiments of the present disclosure, the camera unit 140 may generate a dart target shooting image or a dart target shooting training image under the control of the controller 100 .

The network communication unit 150 may be configured regardless of its communication mode, such as wired and wireless, and may be configured with various communication networks such as a local area network (LAN) and a wide area network (WAN).

The network communication unit 150 may detect a connection state of a network with another device 2000 or the dart game server 3000 and a transmission/reception speed of the network. However, it is not limited thereto, and traffic information related to the transmission/reception speed of the network may be received from the dart game server 3000.

Data received through the network communication unit 150 may be displayed on the display unit 120 or transmitted to another device 2000 .

For example, when receiving data related to a dart game from another device 2000 or the dart game server 3000, the network communication unit 150 may transmit the received data to another device such as a mobile device. The dart game providing method according to the present disclosure may allow two or more devices to participate in the same dart game. Accordingly, the player of the dart game device 1000 may have to wait until one, two or more players participate in the same dart game. Accordingly, the network communication unit 150 may transmit data related to the dart game to the player's mobile device or the like. In this case, the player may take a break and check whether matching with another player has been made through the mobile device.

According to some embodiments of the present disclosure, the controller 100 may provide an offline dart game to the player of the dart game device 1000 when the number of players participating in the dart game is not satisfied and waiting for another player. there is.

Specifically, the controller 100 may receive an input for participating in a dart game from a player through the user input unit 170 . In this case, the controller 100 may create a dart game or participate in one of a plurality of previously created dart games. However, when the number of players of the other device 2000 is insufficient, the players of the dart game device 1000 may have to wait until matching is made. In this case, the controller 100 may provide an offline dart game to the player of the dart game device 1000 in conjunction with searching for another player so that matching with another player can be made.

According to some embodiments of the present disclosure, when the number of players participating in the dart game is insufficient, a dummy player may be created by the dart game server 3000 . Here, the dummy player may be a virtual player rather than a real player. Also, the dart game server 3000 may allow the generated dummy players to participate in at least one dart game where the number of players is insufficient. As an example, the dummy player may be created based on the play result data of the real player, so that the real player playing with the dummy player plays a similar form to any real player as if a real-time dart game was played with the real opponent player. can be performed. Hereinafter, an example of a method for generating a dummy player by the dart game server 3000 according to the present disclosure will be described with reference to FIG. 10 .

The sensing unit 160 may detect the play of the losing player against the dart target unit 200 . For example, the sensing unit 160 may detect the hit position of the dart pin. The sensing unit 160 may electrically convert a score corresponding to the area where the dart is hit and transmit it to the control unit 100 . In addition, the sensing unit 160 may transmit information about the area where the dart pin is hit to the control unit 100, and the control unit 100 is based on the information on the hit position of the dart pin obtained from the sensing unit 160. score can be calculated.

For example, the sensing unit 160 may sense the pressure applied to the dart target by the thrown dart pin. In this case, information indicating that the sensing unit 160 has sensed the pressure applied to the dart target by the thrown dart pin may be transmitted to the control unit 100 . In addition, the controller 100 may control the dart target to be photographed by the camera unit 140 in order to detect the hit position of the dart pin. The controller 100 may obtain information on the hit location of the dart pin using the captured image, and calculate a score based on the information on the hit location of the dart pin obtained using the captured image. .

The user input unit 170 may receive a user's input for controlling the dart game device 1000 .

For example, referring to (b) of FIG. 3 , the user input unit 170 may be implemented as at least one of a keypad, a dome switch, a touch pad (static pressure/static), a jog wheel, and a jog switch. However, it is not limited thereto.

Also, the user input unit 170 may include a short-range communication unit (not shown). When the user input unit 170 includes the short-range communication unit of the network communication unit 150, the user input unit 170 may be configured to receive a user input input by an external console device. As a short range communication technology, Bluetooth, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, or the like may be used.

The lighting unit 180 may be disposed in various parts of the dart game device 1000 to deliver visual effects to the player of the dart game device 1000.

For example, referring to (a) of FIG. 3 , the first lighting unit 180-1 may be formed to extend in a vertical direction along the protrusion in the front direction of the body structure 300. Also, the second lighting unit 180-2 may be disposed below the front surface of the body structure 300. Also, the first lighting unit 180-1 and the second lighting unit 180-2 may output a signal for notifying occurrence of an event of the dart game device 1000. Examples of events generated in the dart game device 1000 may include identification of a dart game player, hit of a dart, change of a dart game player, game end, and the like.

In addition, the lighting units 180-1 and 180-2 may include Light Emission Diodes (LEDs), so that the occurrence of an event may be notified to the user through blinking of the LEDs. Additionally, the lighting unit 180 may change and output the shape of light, the intensity of light, or the blinking cycle according to the position where the dart pin reaches the dart target part.

Additionally, the lighting units 180 - 1 and 180 - 2 may output lighting in a preset pattern in conjunction with the acceptance of the player's login request determined by the controller 100 . Here, the lighting of the preset pattern may be determined in advance by the player of the dart game device 1000.

Referring back to FIG. 2 , the audio output unit 190 may output audio data stored in the storage unit 110 , such as game sound effects, game operation guidance, and game method descriptions. Also, the sound output unit 190 may output a sound signal related to a function (eg, game sound effect) performed by the dart game device 1000 . The sound output unit 190 may include a receiver, a speaker, a buzzer, and the like. Additionally, the sound output unit 190 may vary and output the volume/type of music according to the position where the dart pin reaches the dart target.

In the present disclosure, the sound output unit 190 is a sound, music, or voice corresponding to play-related data among a plurality of pre-stored sounds, music, or voices, based on the player's dart game play-related data of the dart game device 1000. can output

Specifically, when the controller 100 of the dart game device 1000 accepts the player's login request, it may recognize the player's previous play-related data. The controller 100 may control the sound output unit 190 to output a preset sound based on previous play-related data. Alternatively, when accepting the player's login request, the controller 100 may determine whether a pre-stored sound, music, or voice message exists based on the identification information of the player. And, when it is determined that at least one of pre-stored sound, music, or voice message exists, the controller 100 may control the sound output unit 190 to output a corresponding sound.

Additionally, when the player's login is completed, the controller 100 of the dart game device 1000 may generate a personalized guide message to the player based on history information related to the player's dart game play result. For example, if the previous play time of the player is a week ago, the controller 100 may generate a guide message saying "You logged in after a week. Nice to meet you." As another example, if the player has a history of winning the dart game tournament, the controller 100 may generate a guide message saying "Welcome to the winner of the XX tournament". As such, the controller 100 may provide a more entertaining effect to the player by generating a personalized message suitable for the player from the player's history information.

The dart target unit 200 has a bull's eye in the center, and is divided into concentric circles centered on the bull's eye and straight lines extending radially from the bull's eye, and areas (segments) to which individual scores are assigned Score exist Boards may be included. A plurality of accommodating grooves (bits or holes) into which tips of dart pins can be inserted may be formed on the scoreboard. In this case, the arrangement of scores of the dart target unit 200 and the shapes of areas to which scores are assigned may be variably changed. Also, the dart target unit 200 may be implemented in the form of a touch screen.

The body structure 300 may extend in a direction perpendicular to the ground to form the appearance of the dart game device 1000. In addition, the body structure 300 may include the aforementioned display unit 120, the user input unit 170, the dart target unit 200, and other components. However, it is not limited thereto.

Information related to the dart game may be displayed on the display unit 120 provided on the front of the body structure 300 . Here, the information related to the dart game may include a play image of another player, an image generated based on play-related data of another player, or information related to a play time restriction.

Hereinafter, an example of a method for generating information related to a dart game by the control unit 100 of the dart game device 1000 according to the present disclosure will be described through FIGS. 4 to 7 .

4 is a flowchart for explaining an example of a method of identifying a dart pin hit location performed by a dart game device according to some embodiments of the present disclosure.

In some embodiments of the present disclosure, the method for identifying a location of a dart pin hit may include acquiring at least two dart target shooting images in response to dart pin throwing (s100). Here, the dart target shooting image may include an image of a dart target and at least one dart pin hit by the dart target.

Specifically, the controller 100 may recognize that an event in which a dart pin is thrown by a player has occurred. For example, as described above, the sensing unit 160 may sense the pressure applied to the dart target by the thrown dart pin. In this case, information indicating that the sensing unit 160 has sensed the pressure applied to the dart target by the thrown dart pin may be transmitted to the control unit 100 . As another example, the controller 100 may recognize that an event in which the player throws a dart pin has occurred through the camera unit 140 . The controller 100 may control the dart target to be photographed by the camera unit 140 in order to detect the hit position of the dart pin. In this case, the dart target shooting image may include an image of a dart target and at least one dart pin hit by the dart target.

According to some embodiments of the present disclosure, at least two dart target shooting images may be generated by at least two cameras positioned in predetermined directions. In this case, the number of the at least two dart target shooting images may correspond to the number of the at least two cameras.

In detail, when a dart pin is thrown, a plurality of dart target captured images are provided so that the control unit 100 can more accurately identify the hit position of the dart pin using the plurality of dart target captured images. It can be obtained using a camera. For example, the camera unit 140 may be a plurality of cameras, and in this case, the dart target photographing images may be a plurality of images of dart targets photographed in different directions. For example, the camera unit 140 may include two cameras installed on the left and right sides of the top of the dart game device. In this case, the controller 100 may acquire a dart target shooting image in which a camera located on the left side of the upper end of the dart game device is shooting in a direction toward the dart target. In addition, the controller 100 may obtain a dart target shooting image in which a camera located on the right side of the upper end of the dart game device is shooting in a direction toward the dart target. However, it is not limited thereto, and the location and shooting direction of the camera may vary.

The number of dart target shooting images may correspond to the number of cameras. For example, when the number of cameras is two, the number of dart target shooting images may be two. As another example, the number of dart target shooting images may be an integer multiple of the number of cameras. To be specific, for example, in order to prevent a case in which only images in which dart targets and dart pins cannot be identified are obtained as lighting conditions around the dart game device frequently change, a plurality of cameras are installed at regular time intervals. The dart target can be shot n times. In this case, the number of dart shooting images may be n times the number of cameras. However, it is not limited thereto, and various numbers of dart target shooting images may be generated by various methods.

In some embodiments of the present disclosure, the dart pin hit position identification method includes generating hit position information of the at least one dart pin based on at least two dart target shooting images by a dart pin position identification network model ( s200) may be included.

Specifically, when acquiring the dart target captured image, the controller 100 may generate hit location information of the dart pin identified in the dart target captured image by using the dart pin location identification network model. In some examples, the hit location information of the dart pin may include segment location information. For example, the hit location information of the dart pin may include information indicating a location (or a score area) of a segment where the dart pin is hit. In some examples, the hit position information of the dart pin may include information indicating a position of a bit (or receiving groove or hole) in a segment where the dart pin is hit. However, it is not limited thereto, and the hit location information may include various pieces of information.

The dart pin position identification network model may be implemented using various image processing algorithms capable of generating hit position information of a dart pin from a dart target photographed image. For example, the dart pin position identification network model obtains a bounding box image capable of identifying the position of a dart pin from a dart target shooting image, and processes the obtained bounding box image to obtain hit position information of the dart pin. You can include submodels that you create. Additionally, the dart pin location identification network model may include a sub-model for determining hit location information of a dart pin most recently hit when hit location information of a plurality of dart pins is obtained. However, it is not limited thereto, and the dart pin location identification network model may be implemented in various ways.

Hereinafter, an exemplary operation of the dart pin location identification network model for generating hit location information of a dart pin by processing two or more dart target shooting images is described.

In some embodiments of the present disclosure, generating hit location information of a dart pin based on at least two dart target captured images (s200) performs a merging operation on the at least two dart target captured images to obtain input data. It may include generating and generating hit location information of at least one dart pin by processing the input data by the dart pin location identification network model.

As described above, a plurality of dart target shooting images may be obtained using a plurality of cameras so that the control unit 100 can more accurately identify the hit position of the dart pin using the plurality of dart target shooting images. For example, the controller 100 may obtain two dart target shooting images using two cameras by recognizing the throwing of the dart pin.

In some examples, the controller 100 may generate input data capable of processing two dart target shooting images by a dart pin location identification network model. For example, the controller 100 may generate input data by performing a merging operation on at least two dart target captured images. Specifically, for example, the controller 100 may acquire two bounding box images respectively corresponding to two dart target shooting images. Here, various object detection algorithms may be used as an algorithm for obtaining a bounding box image. In some examples, various images for object detection replacing bounding box images may also be used. Also, the controller 100 may merge the two bounding box images into input data that is a single image. As another example, the controller 100 may merge two dart target shooting images into one image. Also, the controller 100 may obtain input data including two bounding box images from one image. However, the present invention is not limited thereto, and the control unit 100 may generate input data in various ways.

In some embodiments of the present disclosure, input data may be an image in which at least two bounding box images are arranged in a predetermined direction. For example, the input data may be an image in which two bounding box images are arranged vertically. However, it is not limited thereto, and the input data may be an image including a bounding box image in various ways.

The dart pin location identification network model may generate hit location information of the dart pin by processing input data. As described above, the hit location information of the dart pin may include segment location information hit by the dart pin. Also, the hit position information of the dart pin may include information indicating a position of a bit within a segment hit by the dart pin. It is not limited thereto, and the hit location information of the dart pin may be various.

The dart pin location identification network model is a network model learned using a training dataset including training image data including at least two dart target shooting images or at least two bounding box images, and a label corresponding to the training image data. can Here, the label may include information about a position of a segment hit by a dart pin identified from the training image data or a position of a bit on the segment. In this case, the dart pin position identification network model may output hit position information including a segment position hit by the dart pin or a bit position of the segment by processing input data based on at least two dart target shooting images. However, it is not limited thereto, and the dart pin location identification network model may be learned using various training datasets.

In some embodiments of the present disclosure, generating hit location information of at least one dart pin based on at least two dart target shooting images by a dart pin location identification network model (s200) may include the above for noise removal. A step of performing pre-processing on the dart target captured image may be included.

In detail, since the dart game device 1000 can be installed in various surrounding environments, a dart target shooting image including a large number of noises caused by too weak or irregular lighting in shooting a dart target can be obtained. In order to prevent this problem, the controller 100 may perform preprocessing for removing noise. For example, preprocessing for removing noise may include image filtering. For example, when an image expresses a frequency band, a high frequency generally appears in a place where there is a lot of change in brightness (eg, a boundary area), and a general background may appear in a low frequency. In the case of removing high frequencies of an image, a blur effect may occur. When the low frequency of the image is removed, an effect of confirming the area of the object appearing on the image occurs. Accordingly, noise removal and blur processing may be performed using a low-pass filter (LPF) and a high-pass filter (HPF). Additionally, in order to prevent the blur effect from blurring the boundary line, Gaussian blur processing may be efficiently performed while maintaining the boundary line by using a bilateral filtering.

As another example, preprocessing for removing noise may include Gaussian filtering. A box filter uses a kernel composed of the same values, but a Gaussian filter can apply a kernel using a Gaussian function. The value of the kernel matrix may be mathematically generated and applied through a Gaussian function. However, it is not limited thereto, and the pretreatment may be performed in various ways.

The dart pin position identification network model may generate hit position information of a plurality of dart pins from a dart target photographed image including a plurality of dart pins. In this case, the controller 100 may determine the hit position information of the dart pin most recently hit among the hit position information of the plurality of dart pins. A detailed embodiment thereof will be described below with reference to FIG. 5 .

5 is another flowchart for explaining an example of a method of identifying a dart pin hit location performed by a dart game device according to some embodiments of the present disclosure.

Since the dart target shooting image in which a plurality of dart pins hit the dart target does not have information about the hit order, the dart game device of the present disclosure uses not only the dart target shooting image but also additional information to hit the dart most recently. The pin's hit location information can be identified. Hereinafter, specific steps of the dart pin hit position identification method using history information will be described with reference to FIG. 5 .

In some embodiments of the present disclosure, the method for identifying a dart pin hit location may include generating first history information in response to a first dart pin throwing (s1100). Here, the first history information may include at least one of a score value determined by the first dart pin throwing or hit location information of at least one dart pin hit the dart target.

The first history information may be used to identify a hit order of some of the plurality of dart pins appearing in the dart target shooting image. For example, when the first history information is generated in response to the first dart pin throwing, the first history information determines hit location information of a dart pin hit by the second dart pin throwing from among a plurality of hit location information. can be used to Here, the first dart pin throwing may refer to dart pin throwing preceding the second dart pin throwing.

In some examples, the first history information may include a score value determined by the first dart pin throwing. In this case, the point value determined by the first dart pin throwing is compared with the point value determined by the second dart pin throwing to provide information capable of identifying the dart pin hit by the second dart pin throwing. can

In some examples, the first history information may include hit location information of a dart pin hit on a dart target. In this case, the dart pin hit by the second dart pin throwing can be identified by excluding the hit position information of the dart pin hit the dart target included in the first history information from among the plurality of hit position information. However, it is not limited thereto, and the first history information may include various types of information.

In some embodiments of the present disclosure, the method for identifying a location of a dart pin hit may include obtaining a dart target shooting image in response to the second dart pin throwing (s1200). Here, the dart target shooting image may include images of a dart target and a plurality of dart pins hit by the dart target.

As described in FIG. 4 , the controller 100 may recognize that the second dart pin throwing has occurred by the player. In this case, the controller 100 may control the dart target to be photographed by the camera unit 140 in order to detect the hit position of the dart pin. Since it is after the first dart pin throwing and the second dart pin throwing have occurred, the dart target shooting image may include images of a dart target and a plurality of dart pins hit by the dart target. In this case, the hit location information of the dart pin obtained from the dart target shooting image may be plural. In this case, as described below, hit location information of the dart pin corresponding to the second dart pin throwing may be identified using the history information.

In some embodiments of the present disclosure, a dart pin hit location identification method is a dart pin hit location corresponding to a second dart pin throwing based on the first history information and a dart target shooting image by a dart pin location identification network model. A step of determining information (s1300) may be included.

As described with reference to FIG. 4 , the controller 100 may generate input data from a captured image of a dart target. The input data may include a bounding box image corresponding to the dart target shooting image. For example, the controller 100 may generate input data including one bounding box image obtained from one dart target image. In some examples, the dart target shooting image may include at least two images each generated by at least two cameras positioned in a predetermined direction. For example, the camera unit 140 may include two cameras installed on the left and right sides of the top of the dart game device. In this case, the controller 100 may acquire a dart target shooting image in which a camera located on the left side of the upper end of the dart game device is shooting in a direction toward the dart target. In addition, the controller 100 may obtain a dart target shooting image in which a camera located on the right side of the upper end of the dart game device is shooting in a direction toward the dart target. However, it is not limited thereto, and the location and shooting direction of the camera may vary.

Exemplarily, when two dart target captured images are obtained, the controller 100 may generate input data by performing a merging operation on the two dart target captured images. Specifically, for example, the controller 100 may acquire two bounding box images respectively corresponding to two dart target shooting images. Here, various object detection algorithms may be used as an algorithm for obtaining a bounding box image. In some examples, the bounding box image may be replaced with one of a variety of images for object detection. Also, the controller 100 may merge the two bounding box images into input data that is a single image. As another example, in a different order, the controller 100 may merge two dart target shooting images into one image. Also, the controller 100 may obtain input data including two bounding box images from one image. However, the present invention is not limited thereto, and the control unit 100 may generate input data in various ways.

According to some embodiments of the present disclosure, the input data includes an image in which at least two bounding box images are disposed in a predetermined direction, and the at least two bounding box images are generated by at least two cameras located in a predetermined direction. may correspond to at least two images.

In some examples, the input data may be an image in which at least two bounding box images are arranged in a predetermined direction. For example, two dart target shooting images may be acquired by at least two cameras positioned in a predetermined direction by throwing the second dart pin. In this case, the input data may be an image obtained by arranging two bounding box images obtained from two dart target shooting images vertically or horizontally. However, it is not limited thereto, and the input data may be an image including a bounding box image in various ways.

When input data is generated, the dart pin position identification network model may generate hit position information of a plurality of dart pins by processing the input data. As described above, the hit location information of the dart pin may include segment location information hit by the dart pin. Also, the hit position information of the dart pin may include information indicating a position of a bit within a segment hit by the dart pin.

Hereinafter, an exemplary dart pin location identification network model for processing a dart target shooting image including a plurality of dart pins hit on the dart target and a detailed operation of the control unit 100 will be described.

In some embodiments of the present disclosure, the step of determining the hit location information of the dart pin corresponding to the second dart pin throwing based on the first history information and the dart target captured image (s1300) is based on the dart target captured image. Generating hit position information of a plurality of dart pins by processing input data by the dart pin position identification network model, and throwing a second dart pin among hit position information of a plurality of dart pins generated based on the first history information It may include determining hit location information of the dart pin corresponding to .

In detail, as described above, the dart pin position identification network model may obtain information on a plurality of hit positions by processing input data related to a captured image of a dart target including a plurality of dart pins hit on the dart target. In this case, the controller 100 may use the first history information to determine hit location information of the dart pin hit by the second dart pin throwing from among a plurality of hit location information. For example, when the first history information includes a point value determined by throwing a first dart pin, the controller 100 determines a point value determined by throwing a first dart pin by throwing a second dart pin. The dart pin hit by the second dart pin throwing may be identified by comparing with the score value. As another example, the controller 100 can identify the dart pin hit by the second dart pin throwing by excluding the hit position information of the dart pin hit the dart target included in the first history information from among the plurality of hit position information. can However, the controller 100 is not limited thereto, and the control unit 100 may determine hit position information of the dart pin corresponding to the second dart pin throwing using the first history information in various ways.

In some embodiments of the present disclosure, the step of determining the hit location information of the dart pin corresponding to the second dart pin throwing based on the first history information and the dart target shooting image (s1300) includes the first history information Hit location information of the dart pin corresponding to the second dart pin throwing may be generated by processing the input data by the dart pin location identification network model.

In detail, instead of post-processing using the first history information after acquiring a plurality of hit location information using the dart pin location identification network model, the first history information is the dart pin location identification network model of discrete data. can be entered in the form For example, when the dart pin location identification network model is a deep learning model including an input layer, a hidden layer, and an output layer, the first history information may be input as input data to the input layer of the dart pin location identification network model. there is. As another example, the first history information may be input as input data to one of the plurality of hidden layers. In this case, the dart pin location identification network model may output only hit location information of the dart pin corresponding to the second dart pin throwing from among a plurality of hit location information. As another example, the dart pin location identification network model may additionally output information capable of identifying hit location information of the dart pin corresponding to the second dart pin throwing. However, it is not limited thereto, and the dart pin location identification network model may process the first history information as input data in various ways.

According to some embodiments of the present disclosure, the method for identifying a location of a dart pin hit may further include generating second history information in response to the second dart pin throwing.

Similar to the first history information, the second history information may be used to identify hit location information of a dart pin hit last among a plurality of dart pins appearing in the dart target photographed image. For example, when the second history information is generated in response to the second dart pin throwing, the second history information determines hit location information of a dart pin hit by the third dart pin throwing from among a plurality of hit location information. can be used to Here, the third dart pin throwing may refer to dart pin throwing following the second dart pin throwing.

The same as or similar to the first history information, the second history information may include a score value determined by the first dart pin throwing. Also, the second history information may include hit location information of a dart pin hit on a dart target. The second history information may be used to determine hit location information of a dart pin hit by the third dart pin throwing in a manner similar to the first history information as described above. However, it is not limited thereto, and the second history information may include various types of information.

6 is a diagram for explaining a dart target shooting image.

As described above, the dart target shooting image 10 may include an image of a dart target and at least one dart pin hit by the dart target. Referring to FIG. 6A , the control unit 100 provides a dart target shooting image 10a in which a camera located on the left side of the upper end of the dart game device is photographed in a direction toward the dart target in response to dart pin throwing, and an upper end of the dart game device. A dart target shooting image 10b photographed in a direction toward the dart target may be acquired by a camera located on the right side of the dart target.

In this case, the controller 100 may generate input data by performing a merging operation on two dart target captured images. Specifically, for example, the controller 100 may acquire two bounding box images 20a and 20b respectively corresponding to the two dart target shooting images. Here, various object detection algorithms may be used as an algorithm for obtaining a bounding box image. In some examples, various images for object detection replacing bounding box images may also be used. Also, the controller 100 may merge the two bounding box images 20a and 20b into input data 20 that is a single image. As another example, the controller 100 may merge two dart target shooting images 10a and 10b into one image. Also, the controller 100 may obtain input data 20 including two bounding box images 20a and 20b from one image. However, the present invention is not limited thereto, and the control unit 100 may generate input data in various ways.

Referring to FIG. 6B, eight bounding box images 21a to 21h are shown in (a), and the eight bounding box images 21a to 21h are merged into a pair corresponding to one dart pin throwing input data. (22a to 22d) are shown in (b). For example, when the bounding box images corresponding to the first dart pin throwing are the image 21a and the image 21e, the input data obtained by merging the two bounding box images may be the input data 22a. Similarly, when the bounding box images corresponding to the first dart pin throwing are images 21b and 21f, input data obtained by merging the two bounding box images may be input data 22b.

7 is a diagram for explaining an exemplary dart pin location identification model 1100 according to some embodiments of the present disclosure.

Referring to FIG. 7A , the control unit 100 obtains a plurality of hit location information using a dart pin location identification network model and then post-processes the first history information, thereby performing a hit location of the most recently thrown dart pin. information can be determined. For example, the dart pin location identification network may generate hit location information 31 of a plurality of dart pins by processing input data 20 including an image representing two dart pins. Here, the input data may be obtained from two dart target shooting images acquired in response to the second dart pin throwing occurring after the first dart pin throwing. For example, the hit position information 30 of the plurality of dart pins includes first dart pin hit position information 31 corresponding to a first dart pin throwing and second dart pin hit position information corresponding to a second dart pin throwing. (32) may be included. In this case, the hit location information selector 1200 uses the history information 40 to determine the most recent throw (eg, the second dart pin throw) among the hit location information 30 of a plurality of dart pins. Corresponding hit location information 32 can be determined.

In some examples, the history information may include various pieces of information for identifying hit location information of a dart pin hit last among a plurality of dart pins appearing in the dart target shooting image. For example, when the first history information 40 is generated in response to the first dart pin throwing, the first history information is a dart pin hit by the second dart pin throwing among the plurality of hit location information 30. It can be used to determine the hit location information 32 of

In some examples, the first history information 40 generated in response to the first dart pin throwing may include a score value determined by the first dart pin throwing. In this case, the hit location information selector 1200 compares the point value determined by the first dart pin throwing with the point value determined by the second dart pin throwing, thereby obtaining the second hit location information among the plurality of hit location information 30. The dart pin hit location information 32 can be determined.

In some examples, the first history information 40 may include hit location information of a dart pin hit on a dart target. In this case, the hit location information selector 1200 removes the hit location information of the dart pin hit on the dart target included in the first history information from among the plurality of hit location information 30, thereby obtaining the second dart pin hit location. Information 32 can be determined.

The hit location information selector 1200 may be implemented as a simple calculation model or as a network model. However, the hit location information selection unit 1200 is not limited thereto and may be implemented in various ways.

Referring to FIG. 7B , instead of post-processing using the first history information 40 after acquiring a plurality of hit location information using the dart pin location identification network model, the control unit 100 provides the first history information ( 40) into the dart pin location identification network model 1100 in the form of discrete data, it is possible to determine the second dart pin hit location information 32 from among a plurality of hit location information. For example, when the dart pin location identification network model 1100 is a deep learning model including an input layer, a hidden layer, and an output layer, the first history information 40 of the dart pin location identification network model 1100 It can be input to the input layer. As another example, the first history information may be input to one of the plurality of hidden layers. In this case, the dart pin position identification network model 1100 outputs only the second dart pin hit position information 32 from among a plurality of hit position information, or can identify the second dart pin hit position information 32 Additional information can be output.

8 is a flowchart illustrating an example of a method of generating a training dataset for a dart pin position identification method performed by a computing device according to some embodiments of the present disclosure.

In order to more accurately identify the hit location of the dart pin using a plurality of dart target shooting images corresponding to the dart pin throwing, the dart pin location identification network 1100 needs to be learned using a sufficient amount of training dataset. there is. In some examples, the dart target unit 200 may include a bullseye and a score board in which areas (segments) divided by straight lines radially extending from the bullseye and to which individual scores are assigned exist. Each segment may include a plurality of receiving grooves (bits or holes) into which dart tips may be inserted. In this case, the hit position of the dart pin can be determined by identifying the segment position or the bit position on the segment. In particular, since the number of segments and the number of bits on a segment are limited, the positions of dart pins may be specified in a limited number.

A method for generating a training dataset for identifying dart pin positions of the present disclosure includes the number of hit dart pins and hit positions of dart pins on the dart target unit 200 including a scoreboard having a limited number of dart pin positions. While changing, a plurality of dart target shooting training images may be acquired. Through this, the method of generating a training dataset for identifying dart pin locations of the present disclosure can generate a sufficient amount of training dataset. In addition, a sufficient amount of training data set may be generated by acquiring dart target shooting training images in various environments where the dart game device 1000 may be located. The method of generating a training dataset for identifying dart pin locations of the present disclosure may be performed by various computing devices including the dart game device 1000, at least another device 2000, and the dart game server 3000.

Hereinafter, a method of generating a training dataset for learning the dart pin location identification network 1100 will be described in detail.

In some embodiments of the present disclosure, a method for generating a training dataset for identifying a dart pin location includes acquiring a dart target shooting training image in which at least one dart pin hits a segment of a dart target (s2100). ) may be included. Here, each of the plurality of segments included in the dart target may include a predetermined number of bits.

As described with reference to FIGS. 4 to 7 , the dart pin position identification network model 1100 may be implemented using various image processing algorithms capable of generating hit position information of a dart pin from a dart target captured image. In some examples, the dart pin location identification network model may generate hit location information of one or more dart pins by processing input data based on a plurality of dart target shooting images obtained from various angles. However, it is not limited thereto, and the dart pin location identification network model may generate hit location information of one or more dart pins by processing input data based on a captured image of one dart target. Hereinafter, a method of generating a training dataset related to the dart pin location identification network model 1100 using a dart target shooting image composed of two or more images will be described. However, it is not limited thereto, and the method according to the present disclosure may be applied to generate a training dataset related to the dart pin location identification network model 1100 using a single dart target shooting image.

As described above, the dart target may be divided into segments to which individual scores are assigned. In addition, each segment may have a plurality of bits (or holes, receiving grooves) that each dart pin can hit. In this case, the hit position where the dart pin can be hit may be limited to the total number of bits included in the plurality of segments. Accordingly, a sufficient amount of dart target shooting training images can be acquired while changing the position of the hit bit of the dart pin.

In some embodiments of the present disclosure, acquiring a dart target shooting training image in which at least one dart pin hits on one segment of the dart target (s2100) includes a first bit included in the first segment of the dart target Acquiring a first dart target shooting training image hit by a first dart pin on a second bit included in the first segment of the dart target and a second dart target shooting training image hit by the first dart pin on a second bit included in the first segment of the dart target It may include the step of obtaining.

In detail, a plurality of dart target shooting training images may be obtained by changing a bit hit by a dart pin on one segment. For example, in the case where the first segment includes first bits and second bits that are different from each other, first dart target shooting training is performed by shooting a dart target hitting the first dart pin on the first bit of the first segment. An image can be acquired. And, a second dart target shooting training image may be obtained by removing the first dart pin on the first bit of the first segment and photographing a dart target hitting the first dart pin on the second bit of the first segment. there is. Here, the first dart pin may be the same or similar one or more dart pins, and does not necessarily mean one dart pin. Compared to the first dart target shooting training image, the second dart target shooting training image may be an image in which a hit position of one dart pin on one segment is changed.

In some embodiments of the present disclosure, acquiring a dart target shooting training image in which at least one dart pin hits on one segment of the dart target (s2100) includes a third bit included in a second segment of the dart target and acquiring a third dart target shooting training image hit by the first dart pin on the image.

In detail, a plurality of dart target shooting training images may be acquired by changing the position of the segment hit by the dart pin. As in the above example, the first dart target shooting training image may be obtained by shooting a dart target hitting the first dart pin on the first bit of the first segment. In addition, a third dart target shooting training image may be obtained by removing the first dart pin on the first bit of the first segment and photographing a dart target hitting the first dart pin on the third bit of the second segment. there is. In this case, compared to the first dart target shooting training image, the third dart target shooting training image may be an image in which a hit position of one dart pin on the two segments is changed.

In some embodiments of the present disclosure, acquiring a dart target shooting training image in which at least one dart pin hits on one segment of the dart target (s2100) includes a first bit included in the first segment of the dart target The method may further include acquiring a fourth dart target shooting training image in which the first dart pin hits the first dart pin and the second dart pin hits the second bit included in the first segment of the dart target.

In detail, dart target shooting training images for two dart pins hit on two different bits included in one segment may be obtained. For example, a dart target in which the second dart pin hits the second bit included in the first segment of the dart target in a state in which the first dart pin hits the first bit included in the first segment of the dart target A fourth dart target shooting training image may be acquired by shooting. In this case, the fourth dart target shooting training image may be an image obtained by adding a hit position of one dart pin on one segment compared to the first dart target shooting training image.

In some embodiments of the present disclosure, acquiring a dart target shooting training image in which at least one dart pin hits on one segment of the dart target (s2100) includes a first bit included in the first segment of the dart target The method may further include acquiring a fifth dart target shooting training image in which the first dart pin hits the first dart pin and the second dart pin hits the fourth bit included in the first segment of the dart target.

In detail, a dart target shooting training image may be obtained by moving one of two hit dart pins on two different bits included in one segment. For example, as in the above example, in a state where the first dart pin hits the first bit included in the first segment of the dart target, the second dart is placed on the second bit included in the first segment of the dart target. A fourth dart target shooting training image may be acquired by shooting the dart target hit by the pin. In this case, while maintaining the first dart pin as it is, the dart in which the second dart pin hit on the second bit included in the first segment of the dart target is moved to the fourth bit included in the first segment A fifth dart target shooting training image may be acquired by shooting the target. In this case, the fifth dart target shooting training image may be an image in which a hit position of one of the two dart pins within one segment is changed compared to the fourth dart target shooting training image.

In some embodiments of the present disclosure, acquiring a dart target shooting training image in which at least one dart pin hits on one segment of the dart target (s2100) includes a first bit included in the first segment of the dart target and obtaining a sixth dart target shooting training image in which the first dart pin hits the first dart pin and the second dart pin hits the third bit included in the second segment of the dart target.

In detail, dart target shooting training images for two dart pins hit on two different segments can be obtained. In the above example, the first dart target shooting training image may be obtained by shooting a dart target hitting the first dart pin on the first bit of the first segment. In addition, by photographing the dart target hit by the second dart pin on the third bit included in the second segment of the dart target while maintaining the first dart pin on the first bit of the first segment, the sixth A dart target shooting training image may be obtained. In this case, the sixth dart target shooting training image may be an image in which a hit position of one dart pin is added to a different segment compared to the first dart target shooting training image.

In some embodiments of the present disclosure, acquiring a dart target shooting training image in which at least one dart pin hits on one segment of the dart target (s2100) includes a first bit included in the first segment of the dart target and obtaining a seventh dart target shooting training image in which the first dart pin hits the first dart pin and the second dart pin hits the fifth bit included in the second segment of the dart target.

In detail, a dart target shooting training image can be obtained by moving one of two hit dart pins on two different segments. In the above example, while maintaining the first dart pin on the first bit of the first segment, the dart target hit by the second dart pin is photographed on the third bit included in the second segment of the dart target. By doing so, the sixth dart target shooting training image can be obtained. In this case, in a state where the first dart pin is maintained on the first bit of the first segment, the second dart pin is transferred to the fifth bit included in the second segment on the third bit included in the second segment of the dart target. A seventh dart target shooting training image may be acquired by shooting the dart target moved. In this case, the seventh dart target shooting training image may be an image in which a hit position of one dart pin on one segment is changed compared to the sixth dart target shooting training image.

According to some embodiments of the present disclosure, acquiring a first dart target shooting training image hit on a first bit included in the first segment of the dart target includes shooting the first dart target under a first shooting environment condition. acquiring training images; and acquiring the first dart target shooting training image under a second shooting environment condition.

In detail, the surrounding environment where the dart game device 1000 is located may have various environments. In this case, the training dataset may include images generated under various shooting environment conditions so that the dart pin location identification network model 1100 can well process images captured in various surrounding environments.

For example, the first photographing environment condition may be a relatively bright lighting condition. And, the second photographing environment condition may be a state having relatively dark lighting. By shooting without changing the hit position of the dart pin under two different conditions, a plurality of dart target shooting training images in which only the shooting environment conditions are changed may be obtained.

As another example, a dart target shooting training image in which only the shooting environment conditions are changed may be obtained by performing post-processing of different conditions on the dart target shooting training image. For example, by adjusting the brightness of the dart target shooting training images, a plurality of dart target shooting training images in which only shooting environment conditions are changed may be obtained. However, it is not limited thereto, and the dart target shooting training image may be obtained in various ways.

According to some embodiments of the present disclosure, the method of identifying a location of a dart pin hit may include assigning a label corresponding to the acquired dart target shooting training image ( S2200 ). Here, the label may be determined based on the position of the segment hit by the dart pin. Also, a label can be determined based on the segment position and the position of the bit on the segment.

When the dart target shooting training image is obtained, a label corresponding to the dart target shooting training image may be assigned. For example, when a dart target shooting training image includes an image of a dart pin hit on a first bit of a first segment, a label corresponding to the dart target shooting training image may be information indicating the first segment. there is. As another example, the label corresponding to the dart target shooting training image may be the first segment and information indicating a bit position on the first segment. Here, assigning a label may mean an operation of associating and storing a dart target shooting training image and a corresponding label. Since a pair of dart target shooting training images and a plurality of labels are created in this way, a training dataset can be created. However, the label is not limited thereto, and the label corresponding to the dart target shooting training image may be assigned in various ways.

As an additional example, a label corresponding to a dart target shooting training image may include a bounding box image. As an additional example, a label corresponding to the dart target shooting training image may further include history information. In the case of a training dataset in which a label includes history information, it may be used to learn the dart pin location identification network model 1100 exemplarily described in FIG. 7B.

According to some embodiments of the present disclosure, the dart target shooting training image includes at least two images corresponding to one label, and the at least two images corresponding to one label are located in a predetermined direction, respectively. It can be created by two cameras.

As described above, the dart pin location identification network model may generate hit location information of one or more dart pins by processing input data based on a plurality of dart target shooting images obtained from various angles. In this case, the training dataset for learning the dart pin location identification network model may include a plurality of dart target shooting images obtained from various angles and one label corresponding to the plurality of dart target shooting images. For example, a plurality of dart target training shooting images obtained from various angles may be acquired using at least two cameras positioned in predetermined directions. In this case, the training dataset may be suitable for a dart game device including at least two cameras positioned in a predetermined direction. However, it is not limited thereto, and the dart target shooting training image may be generated in various ways.

In some examples, when the dart pin location identification network model 1100 processes input images generated from two or more dart target shooting images, the first to seventh dart target shooting training images are respectively 2 It may be two or more images produced by two cameras. For example, the number of dart target shooting training images may correspond to the number of cameras. In some examples, when the number of cameras is two, the number of dart target shooting training images may be two. As another example, the number of dart target shooting training images may be an integer multiple of the number of cameras. Specifically, for example, a plurality of cameras may photograph the dart target n times at intervals of a predetermined time. In this case, the number of dart shooting training images may be n times the number of cameras. However, it is not limited thereto, and various numbers of dart target shooting training images may be generated by various methods.

9 is a schematic diagram illustrating a network function according to some embodiments of the present disclosure.

Throughout this specification, the terms network model, computational model, neural network, network function, and neural network may be used interchangeably. A neural network may consist of a set of interconnected computational units, which may generally be referred to as nodes. These nodes may also be referred to as neurons. A neural network includes one or more nodes. Nodes (or neurons) constituting neural networks may be interconnected by one or more links.

In a neural network, one or more nodes connected through a link may form a relative relationship of an input node and an output node. The concept of an input node and an output node is relative, and any node in an output node relationship with one node may have an input node relationship with another node, and vice versa. As described above, an input node to output node relationship may be created around a link. More than one output node can be connected to one input node through a link, and vice versa.

In a relationship between an input node and an output node connected through one link, the value of data of the output node may be determined based on data input to the input node. Here, a link interconnecting an input node and an output node may have a weight. The weight may be variable, and may be changed by a user or an algorithm in order to perform a function desired by the neural network. For example, when one or more input nodes are interconnected by respective links to one output node, the output node is set to a link corresponding to values input to input nodes connected to the output node and respective input nodes. An output node value may be determined based on the weight.

As described above, in the neural network, one or more nodes are interconnected through one or more links to form an input node and output node relationship in the neural network. Characteristics of the neural network may be determined according to the number of nodes and links in the neural network, an association between the nodes and links, and a weight value assigned to each link. For example, when there are two neural networks having the same number of nodes and links and different weight values of the links, the two neural networks may be recognized as different from each other.

A neural network may be composed of a set of one or more nodes. A subset of nodes constituting a neural network may constitute a layer. Some of the nodes constituting the neural network may form one layer based on distances from the first input node. For example, a set of nodes having a distance of n from the first input node may constitute n layers. The distance from the first input node may be defined by the minimum number of links that must be passed through to reach the corresponding node from the first input node. However, the definition of such a layer is arbitrary for explanation, and the order of a layer in a neural network may be defined in a method different from the above. For example, a layer of nodes may be defined by a distance from a final output node.

An initial input node may refer to one or more nodes to which data is directly input without going through a link in relation to other nodes among nodes in the neural network. Alternatively, in a relationship between nodes based on a link in a neural network, it may mean nodes that do not have other input nodes connected by a link. Similarly, the final output node may refer to one or more nodes that do not have an output node in relation to other nodes among nodes in the neural network. Also, the hidden node may refer to nodes constituting the neural network other than the first input node and the last output node.

In the neural network according to an embodiment of the present disclosure, the number of nodes in the input layer may be the same as the number of nodes in the output layer, and the number of nodes decreases and then increases again as the number of nodes progresses from the input layer to the hidden layer. can In addition, the neural network according to another embodiment of the present disclosure may be a neural network in which the number of nodes of the input layer may be less than the number of nodes of the output layer and the number of nodes decreases as the number of nodes increases from the input layer to the hidden layer. there is. In addition, the neural network according to another embodiment of the present disclosure is a neural network in which the number of nodes in the input layer may be greater than the number of nodes in the output layer, and the number of nodes increases as the number of nodes increases from the input layer to the hidden layer. can A neural network according to another embodiment of the present disclosure may be a neural network in the form of a combination of the aforementioned neural networks.

A deep neural network (DNN) may refer to a neural network including a plurality of hidden layers in addition to an input layer and an output layer. Deep neural networks can reveal latent structures in data. In other words, it can identify the latent structure of a photo, text, video, sound, or music (e.g., what objects are in the photo, what the content and emotion of the text are, what the content and emotion of the audio are, etc.). . Deep neural networks include convolutional neural networks (CNNs), recurrent neural networks (RNNs), auto encoders, generative adversarial networks (GANs), and restricted boltzmann machines (RBMs). machine), a deep belief network (DBN), a Q network, a U network, a Siamese network, a Generative Adversarial Network (GAN), a transformer, and the like. The description of the deep neural network described above is only an example, and the present disclosure is not limited thereto.

The neural network may be trained using at least one of supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning. Learning of the neural network may be a process of applying knowledge for the neural network to perform a specific operation to the neural network.

A neural network can be trained in a way that minimizes output errors. In the learning of the neural network, the learning data is repeatedly input into the neural network, the output of the neural network for the training data and the error of the target are calculated, and the error of the neural network is transferred from the output layer of the neural network to the input layer in the direction of reducing the error. It is a process of updating the weight of each node of the neural network by backpropagating in the same direction. In the case of teacher learning, the learning data in which the correct answer is labeled is used for each learning data (ie, the labeled learning data), and in the case of comparative teacher learning, the correct answer may not be labeled in each learning data. That is, for example, learning data in the case of teacher learning regarding data classification may be data in which each learning data is labeled with a category. Labeled training data is input to a neural network, and an error may be calculated by comparing an output (category) of the neural network and a label of the training data. As another example, in the case of comparative history learning for data classification, an error may be calculated by comparing input learning data with a neural network output. The calculated error is back-propagated in a reverse direction (ie, from the output layer to the input layer) in the neural network, and the connection weight of each node of each layer of the neural network may be updated according to the back-propagation. The amount of change in the connection weight of each updated node may be determined according to a learning rate. The neural network's computation of input data and backpropagation of errors can constitute a learning cycle (epoch). The learning rate may be applied differently according to the number of iterations of the learning cycle of the neural network. For example, a high learning rate may be used in the early stage of neural network training to increase efficiency by allowing the neural network to quickly obtain a certain level of performance, and a low learning rate may be used in the late stage to increase accuracy.

In order to increase the amount of training data for training the neural network, various data augmentation methods may be used. For example, data augmentation may be performed through two-dimensional transformations such as rotation, scale, shearing, reflection, and translation. In addition, data augmentation may be performed by utilizing noise insertion, color, brightness transformation, and the like.

In neural network learning, generally, training data can be a subset of real data (ie, data to be processed using the trained neural network), and therefore, errors for training data are reduced, but errors for real data are reduced. There may be incremental learning cycles. Overfitting is a phenomenon in which errors for actual data increase due to excessive learning on training data. Overfitting can act as a cause of increasing the error of machine learning algorithms. Various optimization methods can be used to prevent such overfitting. To prevent overfitting, methods such as increasing the training data, regularization, inactivating some nodes in the network during learning, and using a batch normalization layer should be applied. can

10 is a simplified and general schematic diagram of an exemplary computing environment in which embodiments of the invention may be implemented.

Although the present disclosure has been described above as being generally embodied by a computing device, those skilled in the art will understand that the present disclosure may be combined with computer-executable instructions and/or other program modules that may be executed on one or more computers and/or may be implemented in hardware and software. It will be appreciated that it can be implemented as a combination.

Generally, program modules include routines, programs, components, data structures, etc. that perform particular tasks or implement particular abstract data types. In addition, those skilled in the art will understand that the methods of the present disclosure can be applied to single-processor or multiprocessor computer systems, minicomputers, mainframe computers as well as personal computers, handheld computing devices, microprocessor-based or programmable consumer electronics, and the like ( It will be appreciated that each of these may be implemented with other computer system configurations, including those that may be operative in connection with one or more associated devices.

The described embodiments of the present disclosure may also be practiced in distributed computing environments where certain tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

Computers typically include a variety of computer readable media. Computer readable media can be any medium that can be accessed by a computer, including volatile and nonvolatile media, transitory and non-transitory media, removable and non-transitory media. Includes removable media. By way of example, and not limitation, computer readable media may include computer readable storage media and computer readable transmission media. Computer readable storage media are volatile and nonvolatile media, transitory and non-transitory, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. includes media Computer readable storage media may include RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital video disk (DVD) or other optical disk storage device, magnetic cassette, magnetic tape, magnetic disk storage device or other magnetic storage device. device, or any other medium that can be accessed by a computer and used to store desired information.

A computer readable transmission medium typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism. Including all information delivery media. The term modulated data signal means a signal that has one or more of its characteristics set or changed so as to encode information within the signal. By way of example, and not limitation, computer readable transmission media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above are also intended to be included within the scope of computer readable transmission media.

An exemplary environment 4100 implementing various aspects of the present disclosure is shown including a computer 4102, which includes a processing unit 4104, a system memory 4106, and a system bus 4108. do. System bus 4108 couples system components, including but not limited to system memory 4106, to processing unit 4104. Processing unit 4104 may be any of a variety of commercially available processors. Dual processor and other multiprocessor architectures may also be used as the processing unit 4104.

The system bus 4108 may be any of several types of bus structures that may additionally be interconnected to a memory bus, a peripheral bus, and a local bus using any of a variety of commercial bus architectures. System memory 4106 includes read only memory (ROM) 4110 and random access memory (RAM) 4112 . A basic input/output system (BIOS) is stored in non-volatile memory 4110, such as ROM, EPROM, or EEPROM, and is a basic set of information that helps transfer information between components within computer 4102, such as during startup. contains routines. RAM 4112 may also include high-speed RAM, such as static RAM, for caching data.

The computer 4102 also has an internal hard disk drive (HDD) 4114 (e.g., EIDE, SATA) - the internal hard disk drive 4114 can also be configured for external use within a suitable chassis (not shown). Yes—a magnetic floppy disk drive (FDD) 4116 (e.g., to read from or write to removable diskette 4118), and an optical disk drive 4120 (e.g., CD-ROM) for reading disc 4122 or reading from or writing to other high capacity optical media such as DVDs). The hard disk drive 4114, magnetic disk drive 4116, and optical disk drive 4120 are connected to the system bus 4108 by the hard disk drive interface 4124, magnetic disk drive interface 4126, and optical drive interface 4128, respectively. ) can be connected to The interface 4124 for external drive implementation includes at least one or both of Universal Serial Bus (USB) and IEEE 1394 interface technologies.

These drives and their associated computer readable media provide non-volatile storage of data, data structures, computer executable instructions, and the like. In the case of computer 4102, drives and media correspond to storing any data in a suitable digital format. Although the description of computer readable media above refers to HDDs, removable magnetic disks, and removable optical media such as CDs or DVDs, those skilled in the art can use zip drives, magnetic cassettes, flash memory cards, cartridges, etc. It will be appreciated that other tangible media readable by the computer, such as the like, may also be used in the exemplary operating environment and any such media may include computer executable instructions for performing the methods of the present disclosure.

A number of program modules may be stored on the drive and RAM 4112, including an operating system 4130, one or more application programs 4132, other program modules 4134, and program data 4136. All or portions of the operating system, applications, modules and/or data may also be cached in RAM 4112. It will be appreciated that the present disclosure may be implemented in a variety of commercially available operating systems or combinations of operating systems.

A user may enter commands and information into the computer 4102 through one or more wired/wireless input devices, such as a keyboard 4138 and a pointing device such as a mouse 4140. Other input devices (not shown) may include a microphone, IR remote control, joystick, game pad, stylus pen, touch screen, and the like. Although these and other input devices are often connected to the processing unit 4104 through an input device interface 4142 that is connected to the system bus 4108, a parallel port, IEEE 1394 serial port, game port, USB port, IR interface, may be connected by other interfaces such as the like.

A monitor 4144 or other type of display device is also connected to the system bus 4108 through an interface such as a video adapter 4146. In addition to the monitor 4144, computers typically include other peripheral output devices (not shown) such as speakers, printers, and the like.

Computer 4102 may operate in a networked environment using logical connections to one or more remote computers, such as remote computer(s) 4148 via wired and/or wireless communications. Remote computer(s) 4148 may be a workstation, computing device computer, router, personal computer, handheld computer, microprocessor-based entertainment device, peer device, or other common network node, and generally includes includes many or all of the components described for, but for brevity, only memory storage device 4150 is shown. The logical connections shown include wired/wireless connections to a local area network (LAN) 4152 and/or a larger network, such as a wide area network (WAN) 4154 . Such LAN and WAN networking environments are common in offices and corporations and facilitate enterprise-wide computer networks, such as intranets, all of which can be connected to worldwide computer networks, such as the Internet.

When used in a LAN networking environment, computer 4102 is connected to local network 4152 through wired and/or wireless communication network interfaces or adapters 4156. Adapter 4156 may facilitate wired or wireless communications to LAN 4152, which also includes a wireless access point installed therein to communicate with wireless adapter 4156. When used in a WAN networking environment, computer 4102 may include a modem 4158, be connected to a communicating computing device on WAN 4154, or establish communications over WAN 4154, such as over the Internet. have other means. A modem 4158, which may be internal or external and wired or wireless, is connected to the system bus 4108 through a serial port interface 4142. In a networked environment, program modules described for computer 4102, or portions thereof, may be stored on remote memory/storage device 4150. It will be appreciated that the network connections shown are exemplary and other means of establishing a communication link between computers may be used.

Computer 4102 is any wireless device or entity that is deployed and operating in wireless communication, eg, printers, scanners, desktop and/or portable computers, portable data assistants (PDAs), communication satellites, wireless detectable tags associated with It operates to communicate with arbitrary equipment or places and telephones. This includes at least Wi-Fi and Bluetooth wireless technologies. Thus, the communication may be a predefined structure as in conventional networks or simply an ad hoc communication between at least two devices.

Wi-Fi (Wireless Fidelity) makes it possible to connect to the Internet without wires. Wi-Fi is a wireless technology, such as a cell phone, that allows such devices, eg, computers, to transmit and receive data both indoors and outdoors, i.e. anywhere within coverage of a base station. Wi-Fi networks use a radio technology called IEEE 802.11 (a, b, g, etc.) to provide secure, reliable, and high-speed wireless connections. Wi-Fi can be used to connect computers to each other, to the Internet, and to wired networks (using IEEE 802.3 or Ethernet). Wi-Fi networks can operate in the unlicensed 2.4 and 5 GHz radio bands, for example, at 11 Mbps (802.11a) or 54 Mbps (802.11b) data rates, or in products that include both bands (dual band) .

Those skilled in the art will understand that information and signals may be represented using any of a variety of different technologies and techniques. For example, data, instructions, instructions, information, signals, bits, symbols and chips that may be referenced in the above description are voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields s or particles, or any combination thereof.

Meanwhile, according to an embodiment of the present disclosure, a computer readable medium storing a data structure is disclosed.

Data structure can refer to the organization, management, and storage of data that enables efficient access and modification of data. Data structure may refer to the organization of data to solve a specific problem (eg, data retrieval, data storage, data modification in the shortest time). A data structure may be defined as a physical or logical relationship between data elements designed to support a specific data processing function. A logical relationship between data elements may include a connection relationship between user-defined data elements. A physical relationship between data elements may include an actual relationship between data elements physically stored in a computer-readable storage medium (eg, a persistent storage device). The data structure may specifically include a set of data, a relationship between data, and a function or command applicable to the data. Through an effectively designed data structure, a computing device can perform calculations while using minimal resources of the computing device. Specifically, the computing device can increase the efficiency of operation, reading, insertion, deletion, comparison, exchange, and search through an effectively designed data structure.

The data structure can be divided into a linear data structure and a non-linear data structure according to the shape of the data structure. A linear data structure may be a structure in which only one data is connected after one data. Linear data structures may include lists, stacks, queues, and decks. A list may refer to a series of data sets in which order exists internally. The list may include a linked list. A linked list may be a data structure in which data are connected in such a way that each data is connected in a single line with a pointer. In a linked list, a pointer can contain information about connection to the next or previous data. A linked list can be expressed as a singly linked list, a doubly linked list, or a circular linked list depending on the form. A stack can be a data enumeration structure that allows limited access to data. A stack can be a linear data structure in which data can be processed (eg, inserted or deleted) at only one end of the data structure. The data stored in the stack may be a LIFO-Last in First Out (Last in First Out) data structure. A queue is a data listing structure that allows limited access to data, and unlike a stack, it can be a data structure (FIFO-First in First Out) in which data stored later comes out later. A deck can be a data structure that can handle data from either end of the data structure.

The nonlinear data structure may be a structure in which a plurality of data are connected after one data. The non-linear data structure may include a graph data structure. A graph data structure can be defined as a vertex and an edge, and an edge can include a line connecting two different vertices. A graph data structure may include a tree data structure. The tree data structure may be a data structure in which one path connects two different vertices among a plurality of vertices included in the tree. That is, it may be a data structure that does not form a loop in a graph data structure.

Throughout this specification, computational model, neural network, network function, and neural network may be used interchangeably. Hereinafter, a neural network is unified and described. The data structure may include a neural network. And the data structure including the neural network may be stored in a computer readable medium. The data structure including the neural network may also include preprocessed data for processing by the neural network, data input to the neural network, weights of the neural network, hyperparameters of the neural network, data obtained from the neural network, activation function associated with each node or layer of the neural network, and neural network It may include a loss function for learning of . A data structure including a neural network may include any of the components described above. That is, the data structure including the neural network includes preprocessed data for processing by the neural network, data input to the neural network, weights of the neural network, hyperparameters of the neural network, data obtained from the neural network, activation function associated with each node or layer of the neural network, and neural network. It may be configured to include all or any combination thereof, such as a loss function for learning of . In addition to the foregoing configurations, the data structure comprising the neural network may include any other information that determines the characteristics of the neural network. In addition, the data structure may include all types of data used or generated in the computational process of the neural network, but is not limited to the above. A computer readable medium may include a computer readable recording medium and/or a computer readable transmission medium. A neural network may consist of a set of interconnected computational units, which may generally be referred to as nodes. These nodes may also be referred to as neurons. A neural network includes one or more nodes.

The data structure may include data input to the neural network. A data structure including data input to the neural network may be stored in a computer readable medium. Data input to the neural network may include training data input during a neural network learning process and/or input data input to a neural network that has been trained. Data input to the neural network may include pre-processed data and/or data subject to pre-processing. Pre-processing may include a data processing process for inputting data to a neural network. Accordingly, the data structure may include data subject to pre-processing and data generated by pre-processing. The foregoing data structure is only an example, and the present disclosure is not limited thereto.

The data structure may include the weights of the neural network. (In this specification, weights and parameters may be used in the same meaning.) Also, a data structure including weights of a neural network may be stored in a computer readable medium. A neural network may include a plurality of weights. The weight may be variable, and may be changed by a user or an algorithm in order to perform a function desired by the neural network. For example, when one or more input nodes are interconnected by respective links to one output node, the output node is set to a link corresponding to values input to input nodes connected to the output node and respective input nodes. A data value output from an output node may be determined based on the weight. The foregoing data structure is only an example, and the present disclosure is not limited thereto.

As a non-limiting example, the weights may include weights that are varied during neural network training and/or weights for which neural network training has been completed. The variable weight in the neural network learning process may include a weight at the time the learning cycle starts and/or a variable weight during the learning cycle. The weights for which neural network learning has been completed may include weights for which learning cycles have been completed. Accordingly, the data structure including the weights of the neural network may include a data structure including weights that are variable during the neural network learning process and/or weights for which neural network learning is completed. Therefore, it is assumed that the above-described weights and/or combinations of weights are included in the data structure including the weights of the neural network. The foregoing data structure is only an example, and the present disclosure is not limited thereto.

The data structure including the weights of the neural network may be stored in a computer readable storage medium (eg, a memory or a hard disk) after going through a serialization process. Serialization can be the process of converting a data structure into a form that can be stored on the same or another computing device and later reconstructed and used. A computing device may serialize data structures to transmit and receive data over a network. The data structure including the weights of the serialized neural network may be reconstructed on the same computing device or another computing device through deserialization. The data structure including the weights of the neural network is not limited to serialization. Furthermore, the data structure including the weights of the neural network is a data structure for increasing the efficiency of operation while minimizing the resource of the computing device (for example, B-Tree, Trie, m-way search tree, AVL tree, Red-Black Tree). The foregoing is only an example, and the present disclosure is not limited thereto.

The data structure may include hyper-parameters of the neural network. Also, the data structure including the hyperparameters of the neural network may be stored in a computer readable medium. A hyperparameter may be a variable variable by a user. Hyperparameters include, for example, learning rate, cost function, number of learning cycle iterations, weight initialization (eg, setting the range of weight values to be targeted for weight initialization), hidden unit number (eg, the number of hidden layers and the number of nodes in the hidden layer). The foregoing data structure is only an example, and the present disclosure is not limited thereto.

Those skilled in the art will understand that the various illustrative logical blocks, modules, processors, means, circuits, and algorithm steps described in connection with the embodiments disclosed herein are electronic hardware, (for convenience) , may be implemented by various forms of program or design code (referred to herein as software) or a combination of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the particular application and the design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.

Various embodiments presented herein may be implemented as a method, apparatus, or article of manufacture using standard programming and/or engineering techniques. The term article of manufacture includes a computer program, carrier, or media accessible from any computer-readable storage device. For example, computer-readable storage media include magnetic storage devices (eg, hard disks, floppy disks, magnetic strips, etc.), optical disks (eg, CDs, DVDs, etc.), smart cards, and flash memory devices (eg, EEPROM, cards, sticks, key drives, etc.), but are not limited thereto. Additionally, various storage media presented herein include one or more devices and/or other machine-readable media for storing information.

It is to be understood that the specific order or hierarchy of steps in the processes presented is an example of example approaches. Based upon design priorities, it is to be understood that the specific order or hierarchy of steps in the processes may be rearranged within the scope of this disclosure. The accompanying method claims present elements of the various steps in a sample order, but are not meant to be limited to the specific order or hierarchy presented.

The description of the presented embodiments is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art of this disclosure, and the general principles defined herein may be applied to other embodiments without departing from the scope of this disclosure. Thus, the present disclosure is not to be limited to the embodiments presented herein, but is to be interpreted in the widest scope consistent with the principles and novel features presented herein.

Claims (13)

A computer program stored on a computer-readable storage medium, wherein the computer program is a training data set for identifying dart pin positions on a dart target composed of a plurality of segments each having a plurality of bits when executed in one or more processors of a computing device. Provides a method for generating, said method:
Acquiring a dart target shooting training image in which at least one dart pin is differently hit in bit units on at least one segment of the dart target - A plurality of segments included in the dart target each have a predetermined number of bits contains -; and
assigning a label corresponding to the acquired dart target shooting training image - the label includes segment position information and bit position information hit by the at least one dart pin appearing on the acquired dart target shooting training image -;
including,
A computer program stored on a computer readable storage medium.
According to claim 1,
The dart target shooting training image includes at least two images corresponding to one label, and the at least two images corresponding to the one label are generated by at least two cameras each positioned in a predetermined direction,
A computer program stored on a computer readable storage medium.
According to claim 1 or 2,
Acquiring a dart target shooting training image in which at least one dart pin hits on at least one segment of the dart target includes:
obtaining a first dart target shooting training image hit by a first dart pin on a first bit included in a first segment of the dart target; and
obtaining a second dart target shooting training image hit by the first dart pin on a second bit included in the first segment of the dart target;
including,
A computer program stored on a computer readable storage medium.
According to claim 3,
Acquiring a dart target shooting training image in which at least one dart pin hits on at least one segment of the dart target includes:
obtaining a third dart target shooting training image hit by the first dart pin on a third bit included in the second segment of the dart target;
Including more,
A computer program stored on a computer readable storage medium.
According to claim 3,
Acquiring a dart target shooting training image in which at least one dart pin hits on at least one segment of the dart target includes:
4th in which the first dart pin hits the first bit included in the first segment of the dart target and the second dart pin hits the second bit included in the first segment of the dart target Acquiring a dart target shooting training image;
Including more,
A computer program stored on a computer readable storage medium.
According to claim 5,
A fifth dart in which the first dart pin hits the first bit included in the first segment of the dart target and the second dart pin hits the fourth bit included in the first segment of the dart target obtaining a target shooting training image;
Including more,
A computer program stored on a computer readable storage medium.
According to claim 3,
Acquiring a dart target shooting training image in which at least one dart pin hits on at least one segment of the dart target includes:
A sixth dart in which the first dart pin hits the first bit included in the first segment of the dart target and the second dart pin hits the third bit included in the second segment of the dart target obtaining a target shooting training image;
Including more,
A computer program stored on a computer readable storage medium.
According to claim 7,
Seventh in which the first dart pin hits the first bit included in the first segment of the dart target and the second dart pin hits the fifth bit included in the second segment of the dart target Acquiring a dart target shooting training image;
Including more,
A computer program stored on a computer readable storage medium.
According to claim 3,
Acquiring a first dart target shooting training image hit on a first bit included in a first segment of the dart target includes:
acquiring the first dart target shooting training image under a first shooting environment condition; and
acquiring the first dart target shooting training image under a second shooting environment condition;
including,
A computer program stored on a computer readable storage medium.
delete delete A method for generating a training dataset for identifying dart pin positions on a dart target composed of a plurality of segments each having a plurality of bits, performed by a computing device,
obtaining a dart target shooting training image in which at least one dart pin is differently hit in units of bits on at least one segment of the dart target; and
assigning a label corresponding to the acquired dart target shooting training image - the label includes segment position information and bit position information hit by the at least one dart pin appearing on the acquired dart target shooting training image -;
including,
method.
A computing device for generating a training dataset for identifying dart pin positions on a dart target composed of a plurality of segments each having a plurality of bits,
memory containing computer-executable components;
a processor executing the following computer-executable components stored in a memory; including,
the processor,
Acquiring a dart target shooting training image in which at least one dart pin is differently hit in units of bits on at least one segment of the dart target, and
assigning a label corresponding to the acquired dart target shooting training image, wherein the label includes segment position information and bit position information hit by the at least one dart pin appearing on the acquired dart target shooting training image; ,
computing device.

Images